von Willebrand Disease in Hong Kong Chinese Women with Unexplained Menorrhagia

HKJGOM 2013; 13(1) 101
Correspondence to: Dr. SF Yip Email: [email protected]
von Willebrand Disease in Hong Kong Chinese Women with Unexplained Menorrhagia
WK TAM MBBS, MRCOG Department of Obstetrics and Gynaecology, Tuen Mun Hospital, Tuen Mun, Hong Kong LK SIU MSc, MSc (Hons) Medical Laboratory Science Department of Pathology, Tuen Mun Hospital, Tuen Mun, Hong Kong YM LEUNG MBBS, MRCOG, FHKCOG Department of Obstetrics and Gynaecology, Tuen Mun Hospital, Tuen Mun, Hong Kong YT YEUNG BSc, AIMLS, MBA Department of Pathology, Tuen Mun Hospital, Tuen Mun, Hong Kong KM CHOW MBChB CT TAM MBBS WS WU MBChB Department of Obstetrics and Gynaecology, Tuen Mun Hospital, Tuen Mun, Hong Kong SF YIP FRCP, FRCPath Department of Pathology, Tuen Mun Hospital, Tuen Mun, Hong Kong
Objectives: To study the prevalence of von Willebrand disease in Hong Kong Chinese women with menorrhagia and to evaluate the utility of ‘Standard Bleeding Questionnaire’ for the screening of menorrhagic women for von Willebrand disease. Methods: Over an 8-month period, prospective data obtained from patients with unexplained menorrhagia recruited during their first gynaecology clinic visit were analysed. All of them were invited to fill in the ‘Standard Bleeding Questionnaire’ by themselves. In the final data analysis, there were 100 eligible patients. Results: The prevalence of von Willebrand disease in the study population was 6% (type I, n=4; type II, n=2). The sensitivity of the individual questions in the questionnaire ranged from 0 to 100%. Combined analysis of three questions with the most satisfactory results (symptoms affecting daily life, symptoms since menarche, and a history of anaemia) yielded a sensitivity of 83%, specificity of 60%, positive predictive value of 12%, and negative predictive value of 98%. Conclusion: The prevalence of von Willebrand disease in patients with unexplained menorrhagia was higher than that in the general population. Detailed history taking and use of simple screening questions can help identify this high-risk group for further diagnostic testing of von Willebrand disease. Hong Kong J Gynaecol Obstet Midwifery 2013; 13(1):101-7
Keywords: Menorrhagia; Prevalence; Questionnaires; von Willebrand diseases
Introduction Menorrhagia is a common clinical problem. Around 5 to 10% of women of reproductive age seek medical attention for menorrhagia. Uterine fibroids (30%) and polyps (10%) are the commonest causes. Less common causes include endometrial hyperplasia, cancer, or medical disease such as thyroid disorders. However, in more than 50% of subjects an anatomical pathology is not found in the genital tract, and menorrhagia remains unexplained1.
In the past decade, there has been increasing awareness of haemostatic problems as the cause of unexplained menorrhagia in a significant number of patients. Menorrhagia is a common presenting symptom in women
with bleeding disorders, of which von Willebrand disease (VWD) is the commonest inherited bleeding disorder and according to the literature in western populations, its quoted prevalence is reported to be approximately 1%2,3. The condition is due to a qualitative or quantitative deficiency of von Willebrand factor (VWF), which is important in the maintenance of primary haemostasis. There are three major types: type I is a partial quantitative deficiency of a normal VWF and accounts for 70 to 80% of all cases; type II (accounting for 20%) includes several qualitative defects
WK TAM et al
HKJGOM 2013; 13(1)102
affecting its multimeric structure or function; and type III (5-10% of all cases) with complete deficiency of VWF and a secondary severe deficiency of FVIII1,4,5. The patients with VWD have a propensity for mucocutaneous bleeding, menorrhagia, easy bruising, epistasis, and postoperative wound bleeding6-8.
The reported frequency of VWD in women with menorrhagia was 5 to 20%9-14, which was substantially more than that in the general population. A systematic review estimated that the overall prevalence of VWD in women presenting with menorrhagia was 13%15.
Most studies on the prevalence of VWD and bleeding disorders in women with menorrhagia were conducted in the West. There were only two studies in Asians—one from India and one from Taiwan, and both reported a prevalence rate of 16%9. Similar studies have not been conducted in the Hong Kong Chinese population.
The detection of haemostatic disorder in patients with menorrhagia demands a broad panel of specialised laboratory investigations. However screening all females with otherwise unexplained heavy menstrual flow for full haemostatic evaluation might not be economical from the public health perspective. In reality, many patients show normal investigation results, despite comprehensive testing if the disorder to be detected is not prevalent. Therefore, in the West, a cost-effective strategy to screen patients for haemostatic disorder was proposed. In 2008, the American College of Obstetricians and Gynecologists (ACOG) advocated the use of a simple questionnaire-based tool to screen females before more elaborate evaluation for bleeding disorders16,17. Screening was considered positive if any one of the following four criteria were met: (1) duration of menses of >7 days and the women reported either flooding or impairment of daily activities with most periods; (2) a history of treatment for anaemia; (3) a family history of a diagnosed bleeding disorder; (4) a history of excessive bleeding with tooth extraction, delivery, miscarriage, or surgery. A combination of eight questions in these four categories demonstrated a high sensitivity of up to 82%17. This questionnaire may be a simple and useful tool to aid practising gynaecologists in Hong Kong for the diagnostic evaluation of patients with menorrhagia.
The purpose of our study was to investigate the prevalence of bleeding disorder and notably VWD in women with menorrhagia in Hong Kong. We set out to study the clinical characteristics of these patients and validate this ‘Standard Bleeding Questionnaire’ (SBQ)
for selecting patients warranting more comprehensive diagnostic testing.
Methods Study Subjects A prospective observational cohort study was recruited from March to December 2011. All the patients were Chinese, aged 18 to 55 years, attending the gynaecological outpatient clinic, Tuen Mun Hospital, Hong Kong. They all had established menorrhagia, based on fulfilling any one of the following criteria4,15,18-20—(1) menses duration of >7 days; (2) heavy flow of >4 days; (3) requirement of pad change in less than 2 hours during a heavy flow period; (4) soaking through bed clothes; and (5) anaemia with a haemoglobin level of <116 g/l.
Patients were excluded from the study if they had menorrhagia: (1) due to uterine pathology noted from physical examination, endometrial aspiration, or pelvic ultrasound (e.g. fibroids, endometrial polyp, endometrial hyperplasia, and endometrial cancer); (2) due to any medical (non-haematological) disease such as hypothyroidism or hyperthyroidism; (3) associated with use of anticoagulants within the past 2 months; (4) associated with use of an intrauterine device; and (5) associated with use of an antiplatelet agent (non-steroidal anti-inflammatory drug, aspirin, or clopidogrel) within the last 14 days.
Consent was obtained from all patients and the study was approved by the hospital institutional review board.
Clinical Assessment Eligible patients were invited to fill in the SBQ (Appendix) by themselves. This questionnaire was a Chinese translation modified from a screening questionnaire published by ACOG in 2008, and included questions also suggested by the National Heart, Lung, and Blood Institute 2008 Guidelines16,17.
A routine gynaecological history was obtained and an examination (abdominal and per-vaginal examination) performed by qualified gynaecologists. Pelvic ultrasound was performed for patients suspected of having a uterine pathology. An endometrial aspirate was obtained for all women aged >40 years, or with other risk factors (according to routine institutional practice).
Questionnaire The questionnaire with 16 questions was given to all study participants. The questions assessed the severity of menstrual bleeding, family history of bleeding disorder,
von Willebrand Disease with Unexplained Menorrhagia
HKJGOM 2013; 13(1) 103
excessive bleeding after specific events (tooth extraction, surgery, or delivery), and a history of anaemia as well as its response to any treatment. Each question required an answer with a pre-coded response; either ‘yes’, ‘no’, or ‘unknown’. The questionnaire also included several ‘contingency questions’, for which a ‘yes’ response elicited further questions. If the answer was ‘no’, such subsequent questions were skipped. All the ‘unknown’ answers were treated as not having that problem.
Laboratory Assessment For all patients, 10 ml of blood was collected for haemostatic studies (complete blood count, prothrombin time [PT] and activated partial thromboplastin time [APTT], fibrinogen, Factor VIII, VWF:Ag, VWF:ristocetin cofactor [RCof] activity). Thyroid function tests were also checked for symptomatic patients. The measurement of RCof activity for VWF was measured by a commercial assay kit (model 700 Aggregometer; Chrono-Log Corporation, USA). VWF antigen assay was measured using a VIDAS VWF kit (Bimerieux Corp, France). Factor VIII assay was measured by a one-stage APTT assay in a Sysmex CA- 7000 Coagulometer (Siemens, Germany).
All patients diagnosed to have VWF abnormalities or VWD were required to have the consistent abnormalities demonstrated in repeat samples. Type I VWD was diagnosed if both VWF:Ag and VWF:RCof were concordantly reduced below the reference range. The diagnosis of type II VWD was established when VWF:RCof level was reduced and the VWF:RCof / VWF:Ag ratio of ≤0.6. Type II VWD abnormality was suggested when the VWF:RCof level was borderline and still within the normal reference range, but
the VWF:RCof / VWF:Ag ratio was consistently reduced to ≤0.6.
Results A total of 108 Chinese women were recruited from the outpatient clinic. Two of these were excluded as they had uterine fibroids (subsequently confirmed by ultrasound), three because endometrial sampling showed abnormal uterine pathology (polyp, complex hyperplasia and carcinoma of corpus), and three others as they defaulted blood taking. The remaining 100 patients who satisfied the entry criteria completed the bleeding questionnaire and had blood testing were included in the analysis.
Patient Characteristics The mean age of the patients was 43.7 (range, 19-55) years, their mean haemoglobin level was 114.2 (range, 72-146) g/l. All the patients had normal platelet counts, fibrinogen levels, PTs, and APTTs. The laboratory characteristics of all patients are summarised in Table 1.
Identification of Patients with Reduced von Willebrand Factor and von Willebrand Disease Six of the 100 patients were confirmed to have consistently abnormal VWF level parameters after repeat testing and diagnosed VWD. Thus, the prevalence of VWD in this study was 6%. Four of these patients manifested type I abnormalities concordant with reduced VWF:RCof and VWF:Ag levels, whereas two patients manifested VWD type II abnormalities with a more significant relative reduction of VWF:RCof than VWF:Ag. The haemostatic results of these six patients are shown in Table 2. In all, four of these six patients (3 type I, 1 type II) had mild anaemia.
Abbreviations: APTT = activated partial thromboplastin time; PT = prothrombin time; VWD = von Willebrand disease; VWF:Ag = von Willebrand factor antigen; VWF:RCof = von Willebrand factor: ristocetin cofactor activity
Table 1. Clinical and laboratory characteristics of VWD and non-VWD patients
Characteristic Mean ± standard deviation p Value All patients (n=100) No bleeding disorders (n=94) VWD (n=6)
Age (years) 43.7 ± 7.2 46.2 ± 4.1 43.6 ± 7.31 0.40
Haemoglobin (g/l) 114.2 ± 16.9 114.4 ± 16.9 112.2 ± 16.5 0.76 Platelet (g/l) 282.5 ± 74.2 283.8 ± 75.4 262.7 ± 50.8 0.50 PT 10.9 ± 0.5 10.9 ± 0.6 10.9 ± 0.4 0.87 APTT 28.7 ± 2.0 28.5 ± 1.9 31.4 ± 1.5 0.004 Fibrinogen (g/l) 2.8 ± 0.6 2.8 ± 0.6 2.7 ± 0.6 0.71 VWF:Ag (%) 98.3 ± 25.6 96.8 ± 31.9 53.3 ± 6.2 0.13 VWF:RCof (%) 94.2 ± 32.7 100.0 ± 24.1 72.5 ± 36.6 0.001 Factor VIII (%) 140.1 ± 47.7 143.0 ± 47.7 94.7 ± 45.7 0.127
WK TAM et al
HKJGOM 2013; 13(1)104
Results Based on Standard Bleeding Questionnaire The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of individual questions are shown in Table 3. The sensitivity of the questions ranged from 0 to 100%. As a group, patients with VWD showed more frequent menorrhagia since menarche and an anaemia history than those without VWF abnormalities. In all, 67% (4/6) of the patients with VWD
had menorrhagia since menarche compared with 35% (33/94) in those without the disease. Similar results were evident for a history of anaemia; with 67% (4/6) in patients with VWD compared with 40% (38/94) in those without the disease. Family history and personal bleeding history (bruising, epistaxis, gastrointestinal, and dental bleeding) did not differ significantly between the groups with and without VWD. Two patients with VWF abnormalities had
Abbreviations: VWD = von Willebrand disease; VWF:Ag = von Willebrand factor antigen; VWF:RCof = von Willebrand factor: ristocetin cofactor activity
Table 2. The data of haemostatic tests of the six patients with VWD
Patient No. Age (years)
Type of VWD
Reference range - 52-200 58-166 50-150 116-155 - - 12 45 134 60 182 106 0.4 II 49 49 50 52 63.4 90 1 I 61 47 54 50 64.4 115 0.9 I 75 39 50 47 96 99 0.9 I 90 51 103 62 98 130 0.6 II 98 46 46 49 68.2 133 1.1 I
Table 3. Statistical characteristics of individual questions in the questionnaire
Ques- tion No.
Question % positive (positive/
% (95% CI) Sensitivity Specificity PPV NPV
1 Menses >7 days 74 (74/100) 67 (24-94) 26 (17-36) 5.4 (1.7-14) 92 (73-99)
2 Flooding sensation 90 (90/100) 100 (52-100) 10.6 (5-19) 6.7 (2.7-14) 100 (66-100) 3 Symptoms affect daily life 60 (60/100) 50 (14-86) 39.4 (30-50) 5 (1.3-15) 92.5 (79-98) 4 Symptoms since menarche 37 (37/100) 67 (24-94) 65 (54-74) 10.8 (4-26) 96.8 (88-99) 5 History of anaemia 42 (42/100) 67 (24-94) 59.6 (49-69) 9.5 (3-24) 97 (76-97) 6 Positive family history 2 (2/100) 0 (0-4.8) 98 (92-100) 0 (0-8) 94 (87-97) 7 Dental surgery (n=55) 7.1 Bleeding after dental surgery 3.6 (2/55) 0 (0-8) 96 (86-99) 0 (0-8) 96 (86-99) 8 General surgery (n=37) 8.1 Excessive bleeding after general surgery 2.7 (1/37) 50 (27-97) 100 (88-100) 100 (5-100) 97 (84-100) 9 Pregnancy (n=82) 9.1 Excessive bleeding after delivery/abortion 3.7 (3/82) 0 (0-5) 96 (88-99) 0 (0-7) 94 (85-98) 10 Epistaxis 4 (4/100) 0 (0-5) 96 (89-99) 0 (0-6) 94 (86-97) 11 Bruises >2 cm 9 (9/100) 0 (0-5) 90 (8-95) 0 (0-4) 93 (86-97) 12 Small wound bleeding >5 mins 4 (4/100) 0 (0-5) 96 (89-99) 0 (0-6) 94 (86-97) 13 Oral/GI bleeding 4 (4/100) 0 (0-5) 96 (89-99) 0 (0-6) 93.8 (86-97) 14 History of transfusion 8 (8/100) 0 (0-5) 92 (83-96) 0 (0-4) 94 (86-97) 15 History of haemorrhagic cyst 4 (4/100) 0 (0-5) 96 (89-99) 0 (0-6) 94 (86-97) 16 Failed treatment 18 (18/100) 0 (0-5) 81 (71-88) 0 (0-2.2) 93 (84-97)
Abbreviations: CI = confidence interval; PPV = positive predictive value; NPV = negative predictive value; GI = gastrointestinal
von Willebrand Disease with Unexplained Menorrhagia
HKJGOM 2013; 13(1) 105
undergone prior surgery, one of whom endured excessive bleeding. Whereas none of the 36 patients without VWF abnormalities who had surgery suffered excessive bleeding.
In view of the wide ranges of sensitivity and specificity for individual questions, we specifically studied the three questions that demonstrated the highest values (Questions 3, 4, and 5). Scoring was assigned based on the number of questions that yielded a positive answer. One point was assigned for each question. Question 8.1 (about excessive bleeding after general surgery) was not included in the analysis, because of the low answer rate (only 2 VWD patients had been challenged), even though that question yielded a high sensitivity (50%) and specificity (100%), and the PPV and NPV rates were 100% and 97%, respectively. The optimal cutoff point for this 3-question scoring system was ≥2 (sensitivity of 83% and specificity of 60%; Table 4).
Discussion Menorrhagia is one of the commonest reasons for referral to the gynaecology outpatient clinic. However, most obstetricians and gynaecologists may not be familiar with the inherited bleeding disorders that give rise to menorrhagia21.
Among these, VWD is the commonest in the western populations, and can account for up to one-fourth of the cases of unexplained menorrhagia. However, it is an under- diagnosed disease in most parts of the world, which could be because of the relatively mild bleeding symptoms in most affected patients, lack of awareness, and the complexity of the laboratory tests involved and their interpretation. In managing patients with menorrhagia, any gynaecological treatment that reduces heavy menstrual bleeding may be appropriate, depending on the woman’s age, gynaecological conditions, and reproductive plans. Medical treatment such as tranexamic acid and oral contraceptive pills remained the first-line treatment. Surgical intervention is sometimes required in patients who cannot tolerate or are unresponsive to medical treatment. Patients with bleeding disorders or VWD deserve special attention and precautions, including
the use of desmopressin (1-desamino-8-D-arginine vasopressin), replacement of deficient factors, family studies, and counselling. Moreover, for patients with platelet function disorders plus VWD, use of non-steroidal anti-inflammatory drugs is discouraged1,22,23.
In addition to menorrhagia, women with VWD are reported to have higher risk of haemorrhagic ovarian cysts and postpartum haemorrhage. Surgical therapy, tranexamic acid, and factor replacement have been used acutely to manage haemorrhagic ovarian cysts8.
In the present study, the prevalence of the VWF abnormalities and VWD in Chinese women with unexplained menorrhagia was 6%. Thus, our VWD prevalence rate was at the lower end of the spectrum cited in the international literature9-14,24.
Study design and its context, and the referral dynamic of the patients can have significant influence on estimations of prevalence. Some of the published studies included patients with menorrhagia due to organic causes (e.g. fibroids)9-12, whereas ours excluded such patients. Our study recruited patients from a general gynaecology outpatient clinic with patients mainly referred by family physicians. We did not recruit inpatients who might have presented late (with more severe symptoms). Moreover, some younger patients were followed up by the paediatricians, not gynaecologists. Some published studies, however, involved patients recruited from more specialised centres or from coagulation clinics9,10,12. Therefore, aside from ethnic differences, the setting of the studies and hence recruitment bias might have accounted for higher prevalence rates reported in some of the studies. The Taiwan study was conducted in a coagulation clinic and observed a VWD rate as high as 16%9. Our study in Asian Chinese appears to be the first study of the genuine prevalence of VWD in general population with menorrhagia.
Our study relied on the patient’s own subjective recall and assessment of the menorrhagia rather than
Abbreviations: CI = confidence interval; PPV = positive predictive value; NPV = negative predictive value
Table 4. Statistical results in combined assessment with questions 3-5
No. of positive questions (Question 3, 4, and 5)
% (95% CI) Sensitivity Specificity PPV NPV
≥1 83 (36-99) 13 (7-22) 6 (2-14) 92 (6-99)
≥2 83 (36-99) 60 (49-69) 12 (4-26) 98 (89-99) ≥3 17 (9-64) 92 (83-96) 11 (1-49) 95 (87-98)
WK TAM et al
References
1. Nichols WL, Hultin MB, James AH, et al. von Willebrand disease (VWD): evidence-based diagnosis and management guidelines, the National Heart, Lung, and Blood Institute (NHLBI) Expert Panel report (USA). Haemophilia 2008; 14:171-232.
2. Rodeghiero F, Castaman G, Dini E. Epidemiological investigation of
the prevalence of von Willebrand’s disease. Blood 1987; 69:454-9. 3. Werner EJ, Broxson EH, Tucker EL, et al. Prevalence of von
Willebrand disease in children: a multiethnic study. J Pediatr 1993; 123:893-8.
4. Kouides PA, Kadir RA. Menorrhagia associated with laboratory
any objective tools. The pictorial blood loss assessment chart and direct measurement by alkaline haematin are the recommended tools to aid the objective documentation of the blood loss in patients with menorrhagia20. In the ACOG study, the pictorial chart enhanced the sensitivity and specificity of the SBQ with respect to identification of VWD. However, these tools are not convenient for patients or easy to be implemented in our local population. Many women in our locality refuse to undertake such measurements. Although our study can be criticised as imprecise, as we relied on memory recall, it more accurately reflects real-life practice and allows better integration of the findings with the everyday management of patients having…